eval_results.json

{
    "epoch": 3.0,
    "eval_/kl_divergence": 59.429229736328125,
    "eval_/mean_score": 0.9832416772842407,
    "eval_loss": 0.0,
    "eval_runtime": 12.1242,
    "eval_samples_per_second": 8.248,
    "eval_steps_per_second": 0.33
}